LESSON 7: SEABORN INTRODUCTION

1. Overall introduction

Seaborn is an open-source Python library for data visualization built-on Matplotlib.

2. Install and import libraries

!conda install -c anaconda seaborn -y

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

3. Using matplotlib with and without seaborn

3.1. Use matplotlib only

import numpy as np
from matplotlib import pyplot as plt

def sin_plot():
    x = np.linspace(-10, 10, 100)
    for i in range(1, 5):
        plt.plot(x, np.sin(x - i))

sin_plot()
plt.show()

3.2. Use matplotlib with seaborn

import numpy as np
from matplotlib import pyplot as plt
import seaborn as sns
sns.set()

def sin_plot():
    x = np.linspace(-10, 10, 100)
    for i in range(1, 5):
        plt.plot(x, np.sin(x - i))

sin_plot()
plt.show()

3.3. Some figure styles in seaborn

import numpy as np
from matplotlib import pyplot as plt
import seaborn as sns
# white, dark, whitegrid, darkgrid, ticks
sns.set_style('darkgrid')

def sin_plot():
    x = np.linspace(-10, 10, 100)
    for i in range(1, 5):
        plt.plot(x, np.sin(x - i))

sin_plot()
plt.show()

3.4. matplotlib and seaborn can be used together

x = np.linspace(-10, 10, 100)
data = pd.DataFrame(data={
    'x': x,
    'sin_x': np.sin(x)
})

sns.set()

# seaborn with plt.title()
sns.lineplot(x='x', y='sin_x', data=data)
plt.title('x and sin(x)')
plt.show()

# seaborn with plt.xlim() and plt.ylim()
sns.lineplot(x='x', y='sin_x', data=data)
plt.xlim(0, 10)
plt.ylim(0, 1)
plt.show()

# seaborn with plt.subplot()
for subplot_index in range(6):
    plt.subplot(3, 2, subplot_index+1)
    sns.lineplot(x='x', y='sin_x', data=data)

plt.tight_layout()
plt.show()

4. Datasets

sns.get_dataset_names()

['anagrams',
 'anscombe',
 'attention',
 'brain_networks',
 'car_crashes',
 'diamonds',
 'dots',
 'exercise',
 'flights',
 'fmri',
 'gammas',
 'geyser',
 'iris',
 'mpg',
 'penguins',
 'planets',
 'taxis',
 'tips',
 'titanic']

data_iris = sns.load_dataset('iris')
data_iris

	sepal_length	sepal_width	petal_length	petal_width	species
0	5.1	3.5	1.4	0.2	setosa
1	4.9	3.0	1.4	0.2	setosa
2	4.7	3.2	1.3	0.2	setosa
3	4.6	3.1	1.5	0.2	setosa
4	5.0	3.6	1.4	0.2	setosa
...	...	...	...	...	...
145	6.7	3.0	5.2	2.3	virginica
146	6.3	2.5	5.0	1.9	virginica
147	6.5	3.0	5.2	2.0	virginica
148	6.2	3.4	5.4	2.3	virginica
149	5.9	3.0	5.1	1.8	virginica

150 rows × 5 columns

data_titanic = sns.load_dataset('titanic')
data_titanic

	survived	pclass	sex	age	sibsp	parch	fare	embarked	class	who	adult_male	deck	embark_town	alive	alone
0	0	3	male	22.0	1	0	7.2500	S	Third	man	True	NaN	Southampton	no	False
1	1	1	female	38.0	1	0	71.2833	C	First	woman	False	C	Cherbourg	yes	False
2	1	3	female	26.0	0	0	7.9250	S	Third	woman	False	NaN	Southampton	yes	True
3	1	1	female	35.0	1	0	53.1000	S	First	woman	False	C	Southampton	yes	False
4	0	3	male	35.0	0	0	8.0500	S	Third	man	True	NaN	Southampton	no	True
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
886	0	2	male	27.0	0	0	13.0000	S	Second	man	True	NaN	Southampton	no	True
887	1	1	female	19.0	0	0	30.0000	S	First	woman	False	B	Southampton	yes	True
888	0	3	female	NaN	1	2	23.4500	S	Third	woman	False	NaN	Southampton	no	False
889	1	1	male	26.0	0	0	30.0000	C	First	man	True	C	Cherbourg	yes	True
890	0	3	male	32.0	0	0	7.7500	Q	Third	man	True	NaN	Queenstown	no	True

891 rows × 15 columns

5. Simple plots

5.1. Line plot

Line plot with single category

sns.lineplot(x='sepal_length', y='sepal_width', data=data_iris)
plt.show()

Line plot with multiple categories

sns.lineplot(x='sepal_length', y='sepal_width', hue='species', data=data_iris)
plt.show()

5.2. Scatter plot

Scatter plot with single category

sns.scatterplot(x='petal_length', y='petal_width', data=data_iris)
plt.show()

Scatter plot customize marker

sns.scatterplot(x='petal_length', y='petal_width', data=data_iris, marker='*')
plt.show()

Scatter plot with multiple categories

sns.scatterplot(x='petal_length', y='petal_width', hue='species', data=data_iris)
plt.show()

Scatter plot with single category with distributions

sns.jointplot(x='petal_length', y='petal_width', data=data_iris)
plt.show()

Scatter plot with multiple categories with distributions

sns.jointplot(x='petal_length', y='petal_width', hue='species', data=data_iris)
plt.show()

Scatter plot with regression curve

sns.regplot(x='sepal_length', y='petal_length', data=data_iris)
plt.show()

Scatter plot with KDE and regression curve

sns.jointplot(x='petal_length', y='petal_width', data=data_iris, kind='reg')
plt.show()

5.3. Bar plot

Bar plot with single category

data_titanic = sns.load_dataset('titanic')
data_titanic

	survived	pclass	sex	age	sibsp	parch	fare	embarked	class	who	adult_male	deck	embark_town	alive	alone
0	0	3	male	22.0	1	0	7.2500	S	Third	man	True	NaN	Southampton	no	False
1	1	1	female	38.0	1	0	71.2833	C	First	woman	False	C	Cherbourg	yes	False
2	1	3	female	26.0	0	0	7.9250	S	Third	woman	False	NaN	Southampton	yes	True
3	1	1	female	35.0	1	0	53.1000	S	First	woman	False	C	Southampton	yes	False
4	0	3	male	35.0	0	0	8.0500	S	Third	man	True	NaN	Southampton	no	True
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
886	0	2	male	27.0	0	0	13.0000	S	Second	man	True	NaN	Southampton	no	True
887	1	1	female	19.0	0	0	30.0000	S	First	woman	False	B	Southampton	yes	True
888	0	3	female	NaN	1	2	23.4500	S	Third	woman	False	NaN	Southampton	no	False
889	1	1	male	26.0	0	0	30.0000	C	First	man	True	C	Cherbourg	yes	True
890	0	3	male	32.0	0	0	7.7500	Q	Third	man	True	NaN	Queenstown	no	True

891 rows × 15 columns

sns.barplot(x='sex', y='survived', data=data_titanic)
plt.show()

Bar plot with multiple categories

sns.barplot(x='sex', y='survived', hue='class', data=data_titanic)
plt.show()

5.4. Histogram

Histogram with kde

kde is Kernel Density Estimation

# kde=False
sns.histplot(x='petal_length', kde=False, data=data_iris)
plt.show()

# kde=True
sns.histplot(x='petal_length', kde=True, data=data_iris)
plt.show()

# kde=True
x = np.random.normal(100, size=100000)
sns.histplot(x, kde=True, bins=100)
plt.show()

Histogram with multiple categories by hue

sns.histplot(x='petal_length', kde=True, data=data_iris)
plt.show()

sns.histplot(x='petal_length', hue='species', kde=True, data=data_iris)
plt.show()

6. Special plots

6.1. Count plot

Count plot with single category

sns.countplot(x='sex', data=data_titanic)
plt.show()

Count plot with multiple categories

sns.countplot(x='sex', hue='class', data=data_titanic)
plt.show()

Count plot horizontally

sns.countplot(y='sex', hue='class', data=data_titanic)
plt.show()

6.2. Box plot

or box and whisker plot

Box plot with single category

sns.boxplot(x='sex', y='age', data=data_titanic)
plt.show()

Box plot with multiple categories

sns.boxplot(x='sex', y='age', hue='class', data=data_titanic)
plt.show()

Show mean on box plot

meanprops = {
    'marker': '+',
    'markeredgecolor': 'white',
    'markersize': '10'
}
sns.boxplot(x='sex', y='age', data=data_titanic, showmeans=True, meanprops=meanprops)
plt.show()

6.3. Violin plot

Mapping between Box plot and Violin plot

Why Violin plot is better than Box plot?

Violin plot with single category

# Box plot
sns.boxplot(x='class', y='age', data=data_titanic)
plt.show()

# Violin plot
sns.violinplot(x='class', y='age', data=data_titanic)
plt.show()

Violin plot with multiple categories

# Box plot
sns.boxplot(x='class', y='age', hue='sex', data=data_titanic)
plt.show()

# Violin plot
sns.violinplot(x='class', y='age', hue='sex', data=data_titanic)
plt.show()

# Violin plot with split=True
sns.violinplot(x='class', y='age', hue='sex', data=data_titanic, split=True)
plt.show()

With more than two categories, we cannot use split=True

sns.boxplot(x='sex', y='age', hue='class', data=data_titanic)
plt.show()

sns.violinplot(x='sex', y='age', hue='class', data=data_titanic)
plt.show()

sns.violinplot(x='sex', y='age', hue='class', data=data_titanic, split=True)
plt.show()

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
/var/folders/9m/8kxkp3s10g9gnjlz2bk_8zs00000gp/T/ipykernel_2751/921161436.py in <module>
----> 1 sns.violinplot(x='sex', y='age', hue='class', data=data_titanic, split=True)
      2 plt.show()

~/opt/anaconda3/envs/work_env/lib/python3.8/site-packages/seaborn/_decorators.py in inner_f(*args, **kwargs)
     44             )
     45         kwargs.update({k: arg for k, arg in zip(sig.parameters, args)})
---> 46         return f(**kwargs)
     47     return inner_f
     48 

~/opt/anaconda3/envs/work_env/lib/python3.8/site-packages/seaborn/categorical.py in violinplot(x, y, hue, data, order, hue_order, bw, cut, scale, scale_hue, gridsize, width, inner, split, dodge, orient, linewidth, color, palette, saturation, ax, **kwargs)
   2385 ):
   2386 
-> 2387     plotter = _ViolinPlotter(x, y, hue, data, order, hue_order,
   2388                              bw, cut, scale, scale_hue, gridsize,
   2389                              width, inner, split, dodge, orient, linewidth,

~/opt/anaconda3/envs/work_env/lib/python3.8/site-packages/seaborn/categorical.py in __init__(self, x, y, hue, data, order, hue_order, bw, cut, scale, scale_hue, gridsize, width, inner, split, dodge, orient, linewidth, color, palette, saturation)
    539         if split and self.hue_names is not None and len(self.hue_names) != 2:
    540             msg = "There must be exactly two hue levels to use `split`.'"
--> 541             raise ValueError(msg)
    542         self.split = split
    543 

ValueError: There must be exactly two hue levels to use `split`.'

6.4. Strip plot

Strip plot is a kind of Scatter plot but it contain one axis as category

Strip plot with single category

sns.stripplot(x='sex', y='age', data=data_titanic)
plt.show()

Strip plot with multiple categories

sns.stripplot(x='sex', y='age', hue='class', data=data_titanic)
plt.show()

6.5. Pair plot

Use to visualize pairwise relationship

data_iris

	sepal_length	sepal_width	petal_length	petal_width	species
0	5.1	3.5	1.4	0.2	setosa
1	4.9	3.0	1.4	0.2	setosa
2	4.7	3.2	1.3	0.2	setosa
3	4.6	3.1	1.5	0.2	setosa
4	5.0	3.6	1.4	0.2	setosa
...	...	...	...	...	...
145	6.7	3.0	5.2	2.3	virginica
146	6.3	2.5	5.0	1.9	virginica
147	6.5	3.0	5.2	2.0	virginica
148	6.2	3.4	5.4	2.3	virginica
149	5.9	3.0	5.1	1.8	virginica

150 rows × 5 columns

sns.pairplot(data_iris,
             hue='species',
             diag_kind='kde',
             kind='scatter',
             palette='husl')
plt.show()

6.6. Heatmap

data_heatmap = np.random.randint(0, 100, (5, 5))
data_heatmap

array([[77, 44,  3, 18, 30],
       [49, 78, 83, 78, 94],
       [99, 42,  0,  4, 38],
       [50, 41, 66, 51, 71],
       [44, 30, 86, 66, 40]])

sns.heatmap(data_heatmap, annot=True)
plt.show()